The proposed research will develop optimization models for the evaluation of four complaints common in ambulatory practice that reduce physician uncertainty at minimal dollar cost. In all other respects the models allow for variability in the values and goals of the physician. The probabilistic estimates required for these models will be a combination of those derived from expert opinion and from statistical analysis of actuarial data obtained in three family practice training centers. In addition, the effectiveness of teaching decision theory and the use of the developed optimization models as tools for education in ambulatory care curricula will be tested.